Mixed Cure Model In Recurrent Event Process And Its Statistical Analysis

With the continuously improvement of technology and medical level, medical statistics in the clinical trials, for some occur repeatedly and is thought to be impossible to cure the disease, in recent years have found suspected cure the existence of the individual. In the process of recurrence, we will consider the effect of cured individuals, promote mixed cure model, to allow more accurate reasonably explain the recurrence process.In the second chapter, the recurrent event data based on the original rate of semiparametric model is based on using Logistic regression to cure rates section,proposed a class of regression model containing healing individuals, to characterize the influence of covariates on the relapse rate. Given the method to estimate the unknown parameters in the model, proves that these large sample properties of the estimate. Numerical simulation and validation of these estimates is also effective in a large data set, and the method is used to set the model and the actual data of bladder cancer.In the third chapter, with the recurrence process, we put forward a mixed cure model of general type, The model contains a kinds of ratio of mixed cure regression model, and the second chapter in the mixed cure model is a special case for it. With the ideas of estimating equations, we give a method of estimating the parameters in the model. Using the theory of modern experience process, we prove that the large sample properties of the estimate.Finally, we will summarize the several studies, and put forward some follow-up research project.